在Rcpp中声明一个变量作为参考 [英] Declare a variable as a reference in Rcpp

问题描述

在C ++中，我们可以声明一个变量作为引用.

In C++, we can declare a variable as a reference.

int a = 10;
int& b = a;

如果我们设置b=15，a也会更改.

If we set b=15, a also changes.

我想在Rcpp中做类似的事情.

I want to do the similar thing in Rcpp.

List X = obj_from_R["X"];
IntegerVector x_i = X[index];
x_i = value;

我想通过向其向量之一插入一个值来从R中更新名为X的对象.上面的代码无法正常工作，因此我尝试了以下方法:

I want to update an object from R called X by inserting a value to one of its vector. The code above did not work, so I tried this:

IntegerVector& x_i = X[index];

并收到错误.

error: non-const lvalue reference to type 'IntegerVector'
      (aka 'Vector<13>') cannot bind to a temporary of type 'Proxy' (aka 'generic_proxy<19>')

推荐答案

这个问题在不同的变体中被问了很多.

This question gets asked a lot in differing variants.

以下是一些受欢迎的答案:

Here are some popular answers:

1. 在C ++函数内，如何将Rcpp对象传递给其他功能(通过引用或复制)?
2. 更新通过引用使用RcppArmadillo submat()通过引用传递的Rcpp :: NumericMatrix
3. Rcpp通过引用与按值传递
4. 通过引用传递data.frame并使用rcpp更新它
5. Rcpp更新矩阵引用并返回R
中的更新
6. 将大型矩阵传递给RcppArmadillo函数而不创建副本(高级构造函数)
7. Rcpp:代理模型的行为不一致
8. 为什么Rcpp实现用于查找唯一项目的数量比基本R慢吗?

1. Within C++ functions, how are Rcpp objects passed to other functions (by reference or by copy)?
2. Update Rcpp::NumericMatrix passed by reference using RcppArmadillo submat()
3. Rcpp pass by reference vs. by value
4. Passing by reference a data.frame and updating it with rcpp
5. Rcpp Update matrix passed by reference and return the update in R
6. Passing large matrices to RcppArmadillo function without creating copy (advanced constructors)
7. Rcpp: Inconsistent behavior with proxy model
8. Why is my Rcpp implementation for finding the number of unique items slower than base R?

---

更多详细信息...

---

More details...

从FAQ条目 Rcpp更改了我按值传递的(const)对象 :

From the FAQ entry Rcpp changed the (const) object I passed by value that I wrote:

Rcpp 对象是基础 R 对象的SEXP或S表达式的包装. SEXP是一个指针变量，用于保存 R 对象数据 R:Internals 的存储位置.也就是说，SEXP不保存 R 对象的实际数据，而仅保存数据所在的引用.为 R 对象创建新的 Rcpp 对象以输入 C ++ 时，该对象将使用与原始 R相同的SEXP 对象(如果类型匹配)，否则必须创建新的SEXP以确保类型安全.本质上，底层SEXP对象是通过引用传递的，而没有显式复制到 C ++ 中.我们将此安排称为代理模型.

Rcpp objects are wrappers around the underlying R objects' SEXP, or S-expression. The SEXP is a pointer variable that holds the location of where the R object data has been stored R:Internals. That is to say, the SEXP does not hold the actual data of the R object but merely a reference to where the data resides. When creating a new Rcpp object for an R object to enter C++, this object will use the same SEXP that powers the original R object if the types match otherwise a new SEXP must be created to be type safe. In essence, the underlying SEXP objects are passed by reference without explicit copies being made into C++. We refer to this arrangement as a proxy model.

因此，&只是视觉上的糖，因为 Rcpp 对象已经充当了引用.

So, the & is just visual sugar as Rcpp objects already act as references.

因此，以下内容将显示结论:

Thus, the following would show the conclusion:

#include <Rcpp.h>

// [[Rcpp::export]]
void show_references(Rcpp::List X, 
                     Rcpp::IntegerVector y, 
                     int index = 0) {
  X[index] = y;
}

示例:

y_vec = c(-1L, 8L, 12L)
X_list = list(a = c(0L, 2L, 3L), b = c(42L, 50L, 30L))
X_list
# $a
# [1] 0 2 3
#
# $b
# [1] 42 50 30

show_references(X_list, y_vec)
X_list
# $a
# [1] -1  8 12
#
# $b
# [1] 42 50 30

我创建的以下 Rcpp 代理模型幻灯片应进一步说明正在发生的情况

The following Rcpp proxy model slides I created should further illustrate what is happening

来源: https://twitter.com/axiomsofxyz/status/938881541396197377

这篇关于在Rcpp中声明一个变量作为参考的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！